The Drunkard's Walk : How Randomness Rules Our Lives Paperback
Leonard Mlodinow's The Drunkard's Walk: How Randomness Rules Our Lives is an exhilarating, eye-opening guide to understanding our random world.
Randomness and uncertainty surround everything we do.
So why are we so bad at understanding them? The same tools that help us understand the random paths of molecules can be applied to the randomness that governs so many aspects of our everyday lives, from winning the lottery to road safety, and reveals the truth about the success of sporting heroes and film stars, and even how to make sense of a blood test.
The Drunkard's Walk reveals the psychological illusions that prevent us understanding everything from stock-picking to wine-tasting - read it, or risk becoming another victim of chance. 'A wonderfully readable guide to how the mathematical laws of randomness affect our lives' Stephen Hawking, author of A Brief History of Time 'Delightful ...Our lives may be shaped by chance, but they are enriched by awareness - just the sort of awareness that this fascinating book will give you' Guardian 'Mlodinow writes in a breezy style, interspersing probabilistic mind-benders with portraits of theorists ...The result is a readable crash course in randomness' The New York Times 'Please read The Drunkard's Walk by Leonard Mlodinow, a history, explanation, and exaltation of probability theory . ..The results are mind-bending' Fortune Leonard Mlodinow has a Ph.D., has been a member of the faculty of the California Institute of Technology and a television writer in Hollywood, as well as developing many award winning CD-ROMs. He is currently Vice President of Emerging Technologies and R&D at Scholastic Inc. and lives in New York City. His previous books include A Brief History of Time, which he co-authored, as well as Euclid's Window and Some Time with Feynman, both published by Penguin.
- Format: Paperback
- Pages: 272 pages, Illustrations
- Publisher: Penguin Books Ltd
- Publication Date: 02/04/2009
- Category: Probability & statistics
- ISBN: 9780141026473
- EPUB from £5.49
Showing 1 - 2 of 2 reviews.
Review by Amtep
I found this a bit heavy on the biographies of dead mathematicians, and a bit light on the surprising effects of probabilities and how they really work. There were some examples of how our intuition can be wrong about probabilities, but I didn't find much that is useful for actually refining my intuition.The point made in the subtitle ("how randomness rules our lives") was barely addressed. The main message was an admonition to judge people on their ability rather than their success or failure. Unfortunately, no tips on how to actually do that.
Review by ElectricRay
The first half of Leonard Mlodinow’s essay on randomness charts the history of chance and gives a primer on the basic principles of probability. This is mildly interesting insofar as it recounts the endeavours of oddball enlightenment characters (many of the French, like Bernouilli family) and how they stumbled on the idea of probability, and iteratively worked out its implications, and hated each other all the while. Mlodinow briefly states and, to my mind, under-emphasises, the historical and cultural significance of probability in the scientific worldview: it affords us an alternative to reductionism: not only is the need to delve ever further into microscopic (and atomic) detail to find “essential” qualities avoided; the importance of cause and effect at all is significantly relegated. No longer must we establish “this caused that”; probability cares simply that “the occurrence of this” is *correlated* with “the occurrence of that”. Old habits die hard, of course, and the reductionist crowd does tend to draw a causative deduction from a strong correlation. The great Enlightenment sceptic David Hume (who sadly does not rates a mention here) would surely spin in his grave. Having armed the reader with some rudimentary knowhow about normal distributions and standard deviations (it is well, and lightly explained: I wish my high school maths teacher had made it this simple), in the second half of the book Mlodinow takes anecdotal pot-shots at the illogicalities of our modern life. They are legion. Some of his examples are fascinating (the probability of *someone* beating Babe Ruth’s record *at some time* are far greater than you’d think) some are a little old hat (the surprisingly high chance of footballers on the same field sharing a birthday), others are enlightening (the same manager night have consistently beaten the January/December annual performance of the Dow, but not the April/March annual performance), the implication being the parameters, periods and frequencies selected to a sample a data set (often chosen after collection of the data) can help to fit the data to a more compelling story) and some are jaw-droppers (in a period, the distribution of a set of mutual fund managers is exactly as you’d expect - some outperformers, some under performers most bunched around the middle: on the other hand, the comparison over two discrete periods was utter chaos: implication: performance across the sector is “volatile” (their word) or “more or less random (“Mlodinow’s”) if this is right, as stated, it is a matter of international outrage.However, I don’t think it is quite the full picture. Firstly, the data set Mlodinow uses to illustrate this point is precisely two. Having spent the first half of the book elegantly explaining how “false positives” - conclusions as to randomness (or non-randomness) should not be drawn from extremely small samples, this looks like a bit of an own goal. Were Mlodinow’s sample 100 periods, and the relative performance between the managers still all over the place, there would be a better case for outrage. Secondly, in lionising the normal distribution, which he says “rules our lives”, Mlodinow skates over the tendency of human interactions – of which are lives are comprised, after all – to be interdependent (that is, not truly random at all). A normal curve plots the distribution of discrete events (i.e., events whose occurrence do not affect each other’s probability). But, if you yell “fire” run for the exit in a cinema, you make it more likely other people will too. If a stock starts falling, its price will drop and people will be inclined to sell. This is why market crashes and bank runs happen. Writers like Phillip Ball and Benoit Mandelbrot have written compellingly about this interdependence. Where there events are reflexive in this way, probabilities follow a “power law” distribution (this looks rather like a normal curve in the middle, but crucially has a much longer, fatter tail at each end). Here, things can and do happen which are so far outside the realms of random probability as to be impossible. These are the “Black Swans” of Nassim Nicholas Taleb’s recent book. The lesson the financial community has learned (well: apparently *hasn’t* learned) is that one models interdependent events as if they were randomly distributed at one’s utter peril: the events accounting for Long Term Capital Management and Lehman Brothers should not have happened in hundreds of millennia.It is a little unwise, therefore, to write off portfolio theory purely on the basis of normal distributions (there are plenty of other grounds on which to write it off). On the other hand, Mlodinow may well be right to suspect that the “wisdom of crowds” aspect from which markets undoubtedly suffer (and markets are well represented by investment managers) conceals a singular lack of differentiation. The interesting exercise, not done, would be to compare the bell curves of individual investment managers. A particular succession of home runs may be put down to the size of a given hitter’s standard deviation, but a batter still has to earn his own bell curve: It may have been unlikely in the extreme for Roger Maris to hit 61 home runs in a season, but it would have been as good as flat out impossible for me to.That said, Mlodinow’s central thesis - that we habitually mistake expertise and talent for a “lucky streak” - is an attractive one and it is interesting exercise to apply it to one’s own sacred cows. And, for that matter, one’s bêtes noires.